Module Chatoyant_provider.LocalSource

Local OpenAI-compatible provider.

Targets local inference servers such as Ollama, LM Studio, mlx-lm/oMLX, llama.cpp server, and vLLM. The public surface stays provider-specific; compatibility quirks are handled internally.

Sourcetype responses_request = Openai.responses_request
Sourcetype responses_response = Openai.responses_response
Sourcetype image_request = Openai.image_request
Sourcetype image_response = Openai.image_response
Sourcetype embedding_request = Openai.embedding_request
Sourcetype embedding_response = Openai.embedding_response
Sourceval chat_request_json : request -> Chatoyant_runtime.Json.t
Sourceval responses_request_json : responses_request -> Chatoyant_runtime.Json.t
Sourceval image_request_json : image_request -> Chatoyant_runtime.Json.t
Sourceval embedding_request_json : embedding_request -> Chatoyant_runtime.Json.t
Sourceval authorization_headers : ?api_key:string -> unit -> (string * string) list
Sourceval chat_response_of_stream_chunks : string list -> (response, string) result
Sourceval response_of_stream_chunks : string list -> (responses_response, string) result